Parkin Dependent Mitophagy Protects Macrophage Against Oxidative Injury And Inammation During Atherogenesis

Mitochondrial oxidative injury induces macrophage inammatory activation and apoptosis during atherogenesis. Timely clearance of dysfunctional mitochondria may therefore be benecial for the survival of macrophages. Based on these principles, our working hypothesis was therefore that mitophagy mediated by the E3 ubiquitin ligase Parkin could have an important role in reducing both oxidative injury and the apoptosis of macrophages under atherogenic conditions. To examine this proposal, in the present study oxidative low-density lipoprotein (ox-LDL) treated THP-1 macrophages were used for the in vitro experiments, and high-cholesterol-fed male apolipoprotein E knockout (ApoE −/− ) mice were used for the in vivo investigations. The results demonstrated that mitophagy was activated both in oxidatively stimulated THP-1 macrophages and in aortic plaque macrophages of high-cholesterol-fed ApoE −/− mice. In ox-LDL treated THP-1 macrophages, both the expression level and mitochondrial translocation of Parkin were increased following oxidative stimulation, whereas silencing Parkin led to impaired mitophagy, which exacerbated macrophage oxidative injury, NF-κB activation and apoptosis. Taken together, these results have demonstrated that mitophagy exerts a protective role in macrophages under atherogenic conditions, and that Parkin is a key mediator in this process.


Introduction
Plaque destabilization and rupture is the primary cause of lethal complications of atherosclerotic cardiovascular disease (AS). The death of macrophages has been demonstrated to be strongly associated with plaque instability and rupture (1). Previous studies have shown that mitochondrial oxidative injury may accelerate macrophage death during atherogenesis (2)(3)(4). Damaged mitochondria may activate diverse cellular pathways, which ultimately leads to macrophage apoptosis (5)(6)(7)(8)(9). Furthermore, overproduction of reactive oxygen species (ROS) due to mitochondria dysfunction may activate the in ammation pathway, which also exacerbates atherosclerosis (6,10,11). The timely elimination of damaged mitochondria may therefore save macrophages from destruction, and alleviate AS.
Damaged mitochondria can be selectively degraded through the autophagy pathway, a process that is termed mitophagy (12). Mitophagy has been reported to play a salutary role in multiple cardiovascular diseases where oxidative stress and mitochondrial damage occur (13)(14)(15). In oxidative low-density lipoprotein (ox-LDL)-treated macrophage, mitophagy induced by melatonin was shown to attenuate IL-1β secretion (16). However, our understanding of mitophagy in the scope of AS is still far from comprehensive (15).
The E3 ubiquitin ligase Parkin has been recognized as an important mediator in mitophagy activation (17,18). Parkin normally resides in the cytosol; whereas, it is translocated to the outer mitochondria membrane (OMM) when mitochondrial depolarization arises (17,19). At the OMM, Parkin may ubiquitinate a subset of membrane proteins, facilitating their recognition and sequestration by the no. 70014; JC-1; Biotium, Inc.), in accordance with the manufacturer's instructions. In cells with a normal MMP, JC-1 forms aggregates that are detected as red uorescence, whereas in cells with impaired MMP, the dye stays in a monomeric form that is detected as green uorescence. The green-to-red uorescence ratio was used as the measure of the MMP level as previously described (22).
Mitochondria mass analysis. Mitochondrial mass loss was measured by a FACS™ technique using Invitrogen MitoTracker™ Deep Red (cat. no. M22426; Thermo Fisher Scienti c, Inc.) staining, as previously described (23). THP-1 macrophages were cultured with 50 nM MitoTracker Deep Red for 15 min at 37°C in the dark. After treatment, cells were trypsinized, washed and resuspended in Gibco Hank's balanced salt solution buffer. Subsequently, 1x10 5 cells were acquired in the FL4 channels of an Accuri C6 ow cytometer (BD Biosciences).
Measurement of cellular ROS. Intracellular ROS were determined using 2',7'-dichloro uorescein diacetate (DCF-DA) (cat. no. D6883; Merck KGaA) as previously described (23). The uorescence intensity of each group was measured using an Accuri C6 ow cytometer (BD Biosciences), and quanti ed as the fold change compared with the baseline.
Measurement of mtROS. mtROS were determined using Molecular Probes® MitoSOX red (cat. no. M36008; Thermo Fisher Scienti c, Inc.) as previously described (24). The uorescence intensity of each group was measured by ow cytometry using an Accuri C6 ow cytometer (BD Biosciences), and quanti ed as the fold change compared with the baseline. Dihydroethidium (DHE). Intracellular ROS were measured using dihydroethidium (DHE) (cat. no. D7008; Merck KGaA) in accordance with the manufacturer's protocol, as described previously (25). The uorescence intensity of each group was measured by ow cytometry (BD Accuri C6) and quanti ed as the fold change compared with the baseline.
Immuno uorescence experiments. Cells cultured on coverslips were xed with ice-cold methanol for 5 min, and then allowed to air-dry for 3 min. The coverslips were washed 3 times, 5 min per wash with PBS. Cells were subsequently blocked for 30  Aortic sinus immunohistochemistry and histology. Cryostat sections (6 µm-thick) of the aortic sinus were cut parallel to the aortic root. Sections were subsequently stained with 200 nM MitoTracker Deep Red for 20 min at room temperature in the dark, washed with 1X PBS, and xed for 5 min in ice-cold acetone.
Confocal microscopy imaging and analysis. Coverslips and tissue sections were imaged using a laser scanning confocal microscope (TCS SP8 STED, 3x magni cation; Leica Microsystems, Ltd.) equipped with a high-resolution AxioCam MRm digital camera (Carl Zeiss, Inc.). Images were deconvolved using Leica Application Suite Advanced Fluorescence software. Colocalization of Parkin or microtubuleassociated protein 1 light chain 3 (LC3) protein (green uorescence) with the mitochondria [mitochondrial cytochrome c oxidase subunit IV (COX IV) or MitoTracker Deep Red; red uorescence] was analyzed and quanti ed using ImageJ software (National Institutes of Health), as previously described (26,27). At least 20 cells per sample were used for the in vitro studies, whereas n = 6 mice from each group were used for the in vivo study.
Preparation of cytosolic and mitochondrial proteins. Cytosolic and mitochondrial proteins of THP-1 macrophages were prepared using the cytosolic/mitochondria fractionation kits (C3601 kit for cells, and C3606 kit for tissues; Beyotime Institute of Biotechnology), in accordance with the manufacturer's instructions, as previously described (23).
Immunoprecipitation. THP-1 cells were harvested and resuspended in 1 ml of lysis buffer (150 mM NaCl, 50 mM Tris-HCl pH 7.6, 1% NP-40, 0.1% SDS, protease inhibitor cocktail (Roche Diagnostics)). Lysates were incubated on ice for 30 min and cleared by centrifugation at 13,000×g for 10 min. Samples were equalized for the protein concentration and incubated with 5 µl of anti-Mfn2 antibody overnight on a rotator. 50µl of washed protein G agarose beads (Roche Diagnostics) were added to the samples. This was followed by incubation for 2 h on a rotator. Next, the beads were pelleted by centrifugation and the supernatant was discarded. The beads were washed three times with lysis buffer followed by resuspension into 2× loading buffer (Invitrogen) and incubation at 95°C for 5 min. After centrifugation, the supernatant was analyzed by Western blotting.
Detection of apoptosis by annexin-V/propidium iodide (AV/PI) dual staining. The THP-1 macrophages were incubated with ox-LDL for the indicated time periods, and subsequently processed with an annexin V-FITC Apoptosis Detection kit (cat. no. 556547; BD Biosciences), according to the manufacturer's instructions. Flow cytometric analysis was performed using an Accuri C6 ow cytometer (BD Biosciences). 5x10 3 events per sample were collected into list mode les and analyzed using FlowJo software (version 7.6.1). The apoptotic rate was determined as the percentage of Q2 + Q4.
Electron microscopy. Electron microscopy was performed using a protocol previously described (28). Observations were made using a JEOL 1400 TEM microscope (JEOL, Ltd.) equipped with a side mount Gatan Orius SC1000 digital camera (Gatan, Inc.). Autophagic vacuoles (AVs) and mitochondria engulfed within AVs were identi ed as previously described (29).
Statistical analysis. All data are presented as the mean ± SEM (n = 3), unless otherwise stated in the Figure legends. Statistical analysis was performed using Student's t-test for two groups, and one-way analysis of variance for multiple groups. All analyses were performed using GraphPad Prism 6.0 (GraphPad Software, Inc.), and P < 0.05 was considered to indicate a statistically signi cant value.

Results
Ox-LDL-induced ROS overproduction and mitochondria damage in THP-1-derived macrophages. Ox-LDL has been recognized as a key athero-relevant inducer of oxidative stress (6). In the present study, we used ox-LDL-treated THP-1 macrophages, a veri ed model to study oxidative injury of mitochondria in macrophages (30,31). Both cellular and mitochondria ROS increased signi cantly as soon as 6 h following ox-LDL (60 µg/ml) treatment, and continued to increase up to 12 h ( Fig. 1A and B). The effects of ox-LDL on mitochondrial membrane integrity were then assessed by monitoring the MMP using the JC-1 dye. After 12 h ox-LDL treatment, the THP-1 macrophages exhibited stronger green uorescence and a higher green-to-red uorescence ratio compared with untreated cells (Fig. 1C), suggesting a decrease in the MMP had occurred following ox-LDL treatment. Another feature of mitochondrial damage is loss of selective permeability of the mitochondria membrane, which, in turn, leads to the release of proapoptotic factors (2,7). Increased cytosol levels of cytochrome c and cleaved caspase-3 were identi ed in ox-LDLtreated macrophages (Fig. 1D). These results indicated that ox-LDL induces mitochondrial damage and ROS overproduction.
Ox-LDL-induced autophagy and mitophagy ux in macrophages. To assess autophagy and mitophagy activity in response to ox-LDL treatment, the expression levels of LC3-II and Beclin-1, both of which are recognized autophagosome markers, were rst measured (32). The conversion of LC3-I into LC3-II is indicative of autophagic activity, and the level of LC3-II present coincides well with the number of autophagosomes. Notably, it was found that ox-LDL treatment led to a marked increase in the levels of LC3-II and Beclin-1 ( Fig. 2A). Functional autophagy denotes the dynamic process of autophagosome formation, cargo sequestration, and eventual lysosomal fusion/degradation, a process that is termed as 'autophagic ux' (33). The level of p62/SQSTM1, an ubiquitin-and LC3-binding adaptor protein that is inversely associated with autophagic ux, was subsequently examined (34). Incubation with ox-LDL led to the reduction of p62, indicating a functional autophagic ux ( Fig. 2A).
In order to better differentiate autophagy induction from downstream inhibition, ba lomycin A1 was used to block the fusion of autophagosomes and lysosomes. The LC3-II level further increased following ba lomycin A1 treatment while p62 accumulated, which implied the presence of functional autophagic ux (Fig. S1).
Consistently, immuno uorescence staining revealed an increase in the numbers of LC3-positive puncta and their co-localization with the mitochondria marker, COX IV ( Fig. 2B and C). Moreover, transmission electron microscopy (TEM) analysis provided direct evidence for mitophagy through observing autophagic structures containing mitochondrion-like cargoes, con rming the formation of mitophagosomes (Fig. 2D). In addition, sequestration of mitochondria into the lysosome compartments was con rmed by co-localization between COX IV and the lysosome marker LAMP-1 (Fig. 2E). Furthermore, mitochondria mass loss was determined by a recently described method with the help of MitoTracker Deep Red staining (Fig. 2F) (35). Together, these results indicated that functional mitophagy ux was induced in response to ox-LDL treatment.
Mitophagy activation in ox-LDL-treated macrophages is dependent on Parkin. The E3 ubiquitin ligase Parkin ful lls an important role in mitophagy activation (18). To explore the role of the Parkin in oxidative stress-induced mitophagy activation, both the level and the cellular location of Parkin were examined. As shown in Fig. 3A, the total level of cellular Parkin showed a slight, but notable, increase at 12 h following ox-LDL treatment. Furthermore, the level of Parkin in the mitochondrial fraction was signi cantly increased, suggesting the translocation of Parkin to the mitochondria (Fig. 3A), which was further evidenced by immuno uorescence staining (Fig. 3B). As previously reported, PINK1 played an import role in recruitment and activation of Parkin after identifying and anchoring on damaged mitochondria (14,15). Accordingly, we found increased PINK1 level in the mitochondrial fraction after ox-LDL treatment in THP-1 macrophages (Fig. 3A). Once activated, Parkin can lead to ubiquitination of a series of mitochondrial membrane proteins, such as Mfn2 (20,21). To test the function of Parkin, we performed immunoprecipitation using an antibody against Mfn2. The resulting immunoprecipitates were analyzed by Western blotting with antibodies against ubiquitin and against Mfn2. Increased ubiquitination level of Mfn2 was demonstrated after ox-LDL incubation, which implied the activation of Parkin (Fig. 3A). It has been reported that polyubiquitination catalyzed by Parkin gives rise to the recruitment of p62 to damaged mitochondria, facilitating their recognition by autophagosomes (20). The levels of p62 and LC3 I/II were then measured in the mitochondria fraction. The results revealed that the levels of p62 and LC3 increased concomitantly with that of Parkin in the mitochondrial fraction following ox-LDL stimulation (Fig. 3A).
Furthermore, to facilitate our analysis of the importance of Parkin in mitophagy, Parkin was silenced, which led to decreased p62 and LC3 levels in the mitochondrial compartment (Fig. 3C). Immuno uorescence staining revealed that knocking down Parkin protein led to a decrease in the extent of mitochondrion-autophagosome co-localization in the macrophages (Fig. 3D and E). Moreover, TEM images revealed a high density of non-functional autophagosomes in Parkin-silenced macrophages, as well as the intracellular accumulation of mitochondrial fragments (Fig. 3F). Taken together, these results suggested that Parkin is essential for mitophagy activation in response to ox-LDL simulation.
Mitophagy impairment by silencing Parkin aggravates macrophage oxidative injury and apoptosis. To further understand the role of mitophagy in ox-LDL-induced oxidative stress, mitophagy was impaired via silencing Parkin. The results revealed that impaired mitophagy resulted in an elevated cellular green-tored JC-1 ratio and an increase in the mitoSOX intensity of THP-1 macrophages both with and without ox-LDL treatment (Fig. 4A and B). Moreover, silencing Parkin further exacerbated the release of cytochrome c and caspase-3 activation in ox-LDL-treated THP-1 macrophages (Fig. 4C). Since both mitochondrial damage and ROS accumulation are known to be strong inducers of apoptosis, macrophage apoptosis was quanti ed by performing AV/PI dual staining. As shown in Fig. 4D and E, 12 h of ox-LDL treatment led to a marked increase in the ratio of AV-positive macrophages, which was increased even further when Parkin was knocked down. Furthermore, mtROS have recently been associated with in ammation via the NF-κB pathway (6). Subsequently, the effect of mitophagy on NF-κB pathway activation in atherogenic macrophages was investigated next. Figure 4F and G show the phosphorylation (Ser-536) and nuclear translocation of RelA (NF-κB p65) after ox-LDL treatment. Silencing of Parkin led to a further increase in NF-κB p65 activation, possibly due to defective mitophagy and mtROS elimination.
Activation of mitophagy in the plaque macrophages of ApoE −/− mice. In order to further validate our ndings in vivo, whether or not mitophagy was also activated in plaque macrophages of high-cholesterol fed ApoE −/− mice was subsequently explored. Aortic root sections of ApoE −/− mice were immunostained with the MOMA-2 antibody to identify macrophages in lesions (Fig. 5A). Both in early (8 weeks) and advanced (16 weeks) lesions, immuno uorescence staining provided clear evidence of autophagosome and mitophagosome formation (Fig. 5B). Mitophagy activation was further con rmed by observation of autophagic structures containing mitochondrion-like cargoes under the TEM microscope (Fig. 5C). Subsequently, whether or not Parkin also participates in mitophagy activation was investigated. As shown in Fig. 5D, mitochondrial translocation of Parkin was also promoted in plaque macrophages of high-cholesterol-fed ApoE −/− mice. Moreover, the levels of both Parkin and p62 in the mitochondrial fraction of the aorta lysis were signi cantly increased upon atherosclerosis development (Fig. 5E).

Discussion
Macrophages occupy a central role during the pathogenesis of atherosclerosis (1). Increasing lines of evidence suggest that mitochondrial oxidative injury, and the subsequent overproduction of mtROS, is closely associated with macrophage death and plaque destabilization (2). Physiologically, mtROS is a natural by-product of the respiratory chain, and performs important signaling functions as an intracellular messenger (36,37). During atherogenesis, progressive mitochondrial damage and respiratory chain dysfunction develops in macrophages in lesions due to severe and prolonged oxidative stress (4,(38)(39)(40). Damaged mitochondria thereby become a major source of cellular ROS, which continuously amplify the oxidative injuries via impairing adjacent mitochondria and exacerbating mtROS production in a feedforward manner (41). In the present study, rapid mitochondria injury and mtROS accumulation were observed in ox-LDL-treated THP-1 macrophages. Furthermore, increased lesional ROS levels were revealed as atherosclerosis progressed in ApoE −/− mice. It is therefore of paramount importance that dysfunctional mitochondria should be eliminated in a timely manner to prevent further damage.
Autophagy is the natural destructive mechanism that orderly degrades and recycles unnecessary or dysfunctional cellular components. In recent years, autophagy and mitophagy have been widely reported in multiple cardiovascular diseases including AS (14,15). Macrophages and foam cells are the predominant component consist the lipid core of atherosclerotic plaque. Macrophage autophagy has already been demonstrated to have an anti-atherogenic effect (42,43). However, the underlying mechanism(s) has yet to be properly elucidated. Previous studies have proposed that potential mechanisms may involve the autophagic removal of damaged mitochondria, in turn reducing ROS injuries (43,44). In the present study, direct evidence has been provided to demonstrate mitophagy activation in oxidative-stimulated macrophages. Damaged mitochondria were shown to be sequestrated and degraded through the autophagosome-lysosome pathway, and the importance of this process was highlighted by an accelerated rate of in ammatory activation and cell apoptosis in macrophages with impaired mitophagy. Vascular smooth muscle cells (VSMCs) are another important cellular component of atherosclerotic plaque and the main component of the brous cap. Autophagy and mitophagy have also been reported in VSMCs. Defect mitophagy in VSMCs is closely related to cell apoptosis and unstable atherosclerotic plaque phenotype (45,46). Therefore, autophagy and mitophagy play an important role in plaque stabilization and may become a potential target for anti-atherogenic therapy.
The E3 ubiquitin ligase Parkin has been recognized as an important mediator in mitophagy activation in a variety of neuromuscular and cardiovascular diseases (47)(48)(49). However, the role of Parkin in atherosclerosis development has yet to be properly examined. In the present study, it has been demonstrated that Parkin is upregulated and translocated to the mitochondria under oxidative conditions, and that the protein is essential for macrophage mitophagy activation both in vitro and in vivo. Knocking down Parkin leads to impaired mitophagy with intracellular retention of mtROS, which promotes macrophage in ammatory activation and cell apoptosis. Possible roles of Parkin and mitophagy during atherogenesis are summarized in Fig. 6. It is worth noting that Parkin-mediated mitochondrial ubiquitination may have different roles in mitochondrial degradation. Upon mitochondrial membrane depolarization, proteasomes are recruited to the mitochondria in a Parkin-dependent manner, leading to the degradation of proteins located at the OMM and intermembrane space, but not in the inner membrane or the matrix (50). Therefore, immunostaining for mitochondrial inner membrane or matrix proteins, such as CypD and COX IV, would be more appropriate for monitoring mitophagy (51).
Interestingly, ROS have been shown to participate in almost all aspects of macrophage-associated atherogenic processes, including oxidative modi cation of LDL, monocyte recruitment, macrophage activation, efferocytosis and extracellular matrix degradation (52)(53)(54). Therefore, in addition to suppressing macrophage in ammatory activation and apoptosis, the anti-atherogenic effect of mitophagy may be involved in all the pathological processes in which ROS participates. Moreover, a link between ROS and autophagy activation has recently been revealed, and recognized as a self-protective mechanism of cells (55)(56)(57). Known mechanisms for ROS-induced autophagy include direct suppression of mammalian target of rapamycin (mTOR) activity and transcriptional regulation of autophagyassociated genes (55,58,59). Therefore, in addition to leading to mitochondrial injury and an accumulation of Parkin, mtROS may be able to directly induce mitophagy through upregulating autophagosome formation. However, those hypotheses need to be con rmed in subsequent studies.

Conclusion
The present study has demonstrated that an accumulation of damaged mitochondria leads to macrophage oxidative injury and atherogenesis. It has been shown that mitophagy serves an important role in eliminating mitochondria and in decelerating the development of atherosclerosis. Mitophagy may therefore provide a promising target for therapeutic intervention in the treatment of atherosclerosis.

Funding
This work was supported by the Natural Science Foundation of China (grant no. 81700399).

Con icts of interest
The authors declare they have no competing interests.

Availability of data and material
The data that support the ndings of this study are available from the corresponding author upon reasonable request.

Code availability
All data, models, or code generated or used during the study are available from the corresponding author by request.

Ethics approval
In vivo protocols were approved by the hospital's Animal Studies Committee (No. 2018-6-36-GZR)   Possible roles of Parkin and mitophagy during atherogenesis.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.